@ChrisHemedinger wrote: Since "date" variables are inferred based on their assigned SAS formats, you could also create a list of these (regardless of name) by examining the FMTINFO function, "Cat" attribute.  Some testing required 😉   %let lib=SASHELP; %let mem=AIR; proc sql noprint; select t1.name into: dateVars separated by ' ' from sashelp.vcolumn t1 where libname="&lib." and memname="&mem" and fmtinfo(scan(t1.format,1),'Cat')="date" ; quit; %put &=dateVars; proc report data=&lib..&mem.; columns _all_; define &datevars. / display style={TAGATTR='format:m/d/yyyy'}; run;   The DEFINE statement is a bit eccentric about the difference between a "SAS Variable List" and a "list of SAS variables".      81 proc report data=&lib..&mem.; SYMBOLGEN: Macro variable LIB resolves to SASHELP SYMBOLGEN: Macro variable MEM resolves to class 82 columns _all_; 83 define height weight / display style={TAGATTR='format:m/d/yyyy'}; ------ 79 ERROR: HEIGHT does not have a numeric suffix. ERROR 79-322: Expecting a -. 84 run;   My SAS does not have SASHELP.AIR; ... View more

Combining the three datasets was the perfect solution! Once combined I was able to create subsets with uncommon ids as well as perform a comparison between the IDs that existed in all three locations. Thank you! ... View more